Analysis Initial Blackout Refrigerants, such as liquid refrigerant22, when allowed to evaporate to atmosphere expand to a gas at several hundred times their liquid volume. These cold vapours, which are heavier than air, displace oxygen, and gravitate to the lowest point in a contained space. In large enough quantities, the result can be a serious oxygen-lean environment. Cold evaporating refrigerant22 gas, for example, is more dense than air and will fill a room from deck level upwards. The volume of the engine room of the Fame was approximately 332cubic metres. The amount of refrigerant in the factory system of the vessel was estimated at between 1701kg and 2835kg. The specific volume of refrigerant22 at atmospheric pressure (101.3kpa) is 274.7dm3/kg. Taking the mean amount of refrigerant as 2268kg, the gas when released would evaporate and expand to approximately 623cubic metres. This would have displaced the air within the engine space and resulted in a refrigerant-22-rich environment. At the time of the occurrence, the vessel's main and auxiliary electrical power was being provided by a shaft alternator powered by the main engine. In the event, the expanded refrigerant acted as an asphyxiant, causing the suffocation of the main engine by displacing oxygen within the engine room. When the main engine shut down, all main as well as auxiliary electrical power was lost. Second Blackout Approximately 16hours after the initial electrical blackout, the engineers were able to start an auxiliary generator located at an upper level in the engine room and re-establish electrical power but, because of low air in their SCBAs, were unable to reach and re-open the previously isolated seawater inlet valves. The vessel suffered a second blackout 30minutes later, when the auxiliary generator stopped. This was due to either: The auxiliary generator ingesting more refrigerant22, which caused it to asphyxiate and shut down; or The lack of seawater to cool the auxiliary generator, causing it to overheat and shut down. Refrigerant Pump Photo3. Port liquid refrigerant pump According to the two engineers who entered the engine room several times after the blackout, the only piece of equipment noted to be damaged was the standby (starboard) refrigerant pump. The pump was reported as having disintegrated or exploded. Although the pump was not in service at the time of the failure, its isolation valves were still open and when the pump failed, the entire charge of refrigerant was free to escape to atmosphere. As the vessel sank, the starboard liquid refrigerant pump was not recovered and the cause for its failure could not be determined. No photograph is available of the pump. Photo3 (taken before the occurrence) shows the similar port liquid refrigerant pump. Factory Deck Overboard Discharge Valves In the absence of any record of damage to the wave gate drain, offal, turbot fishing, and galley waste chutes closing devices or any other structural hull failure affecting the watertight integrity of the vessel, the most likely source of the ingress of seawater onto the factory deck was through one or more of the overboard discharge valves in the starboard side shell plating of the factory tween-deck. After pumping power was lost, it is likely that entrained discarded fish material and other waste debris remaining in the piping system, settled and became lodged in the non-return discharge valves, causing them to malfunction. The absence of anti-syphon loops in the pump discharge piping inboard of the shipside valves, would also have contributed to the ingress of seawater through the valves onto the factory deck. The overboard discharge valves were located above the factory deck and relatively close to the operating waterline such that, while the vessel was in calm water and heeled 3 or 4 to starboard, the valves remained above sea level. This trimmed and heeled condition was maintained for some 13hours after the initial loss of pumping power, and would indicate that the vessel did not ship any significant amount of water throughout that period. As the wind and seas increased, so did the rolling motion of the vessel. When heeled by the wind and rolling approximately 10 to starboard, the valves were at or just below the trimmed waterline (seePhoto4), and became more deeply submerged as the two metre high wave crests passed along the side of the immobile vessel. The repeated immersion of the partially obstructed and ineffective discharge valves, together with the loss of electrical power to the pumps, initiated the continuous and accelerating rate of ingress of seawater through the valves onto the factory deck, which ultimately led to the loss of the vessel. Vessel Sinking Sequence In order to accommodate and facilitate the operation of shrimp processing equipment, the working area extended the full breadth of the vessel and was some 27m in length. The area was not subdivided. Consequently all washing or wastewater and any shipped water was free to flow on the factory deck throughout the working area. In routine operations, catch waste material and washing water from the shrimp catching and processing operations gravitated to the drain wells located on each side of the factory deck and was discharged overboard by the automatic, electrically driven submersible pumps. When electrical power was lost, the ability to discharge wastewater was also lost. Shrimp debris and water, already onboard at that time, remained on the factory deck. Because a certain amount of water was routinely found in the processing factory area, its presence did not cause any undue concern while the tween-deck remained in darkness and largely unattended. As the weather deteriorated, the vessel rolled more due to the increasing seas and was heeled by the wind force, causing all of the existing loose water on the factory deck to gravitate to starboard. The water also flowed toward the after end of the deck because the partially loaded vessel was already trimmed slightly by the stern. The initial slight ingress of water through the partially obstructed starboard-side discharge valves increased when they were more deeply submerged as the immobile and wind-heeled vessel rolled to starboard. The weight and free-surface effects of seawater accumulating on the starboard side of the factory deck increased the angle of heel and reduced the remaining righting ability of the vessel on that side. Shortly before the Fame was abandoned, the level of water at the after end of the factory deck was seen to have reached the battened-down engine room starboard side escape hatch, some 1.67m above the factory deck level and, as the vessel rolled, the flooding also extended across the after end of the factory deck abreast the open engine room door in the port side casing. As the trim by the stern increased, the shipped water downflooded over the coaming of the door in the engine room port side casing and immediately gravitated to the starboard side of the engine room of the heeled vessel. The accumulation of water in the engine room further increased the after trim and angle of heel, and continued until the after starboard quarter of the weather deck became completely immersed (seePhoto2). The additional downflooding through newly immersed deck openings and ventilators, etc. (which had not been effectively sealed to prevent/retard this) at the after end of the weather deck increased the rate at which seawater accumulated in the under-deck compartments at the after end of the vessel. The Fame settled deeper in the water and continued to trim more and more by the after end as the weight of shipped water increased. Such a cycle was maintained until the vessel lost all reserve buoyancy and sank stern first. Immersion Suits The immersion suits were originally designed to be used in an emergency; i.e. when abandoning a vessel. They could be stored and used as required, provided they passed inspection at regular intervals. As the three immersion suits, which were reported to have had problems, were from the same manufacturer, all 10suits were sent back to the factory for inspection and testing. The 11th suit, which had reportedly suffered a zipper failure, was lost with the vessel and could not be examined. At the request of the TSB, a three-step test plan, consisting of a visual inspection, a zipper examination and a leakage test, was carried out by the manufacturer. The visual inspection revealed that all 10suits showed signs of use and that one suit was, in fact, fitted with two right mitts. This non-conformity would not however have precluded the crewmember from using the gloves for thermal protection. The zipper examination yielded no anomaly. Eight of the ten suits failed the leakage test; using a soapy water solution, any detectable leak, no matter how minor, constituted a failure. The leaks ranged from pinhole size leaks in the outer shell to abrasions on the boot. The minor leakage, although not desirable, would have had a negligible effect on the flotation properties or thermal protection of the suit. The larger leaks in way of the scuffs on the boots were most likely caused by the crewmembers during lifeboat and fire drills or the actual abandoning of the vessel. The suit manufacturer advised that the OC 4001model (asused) was later replaced by the OC8001model, which includes more robust, moulded one-piece boots and improved wrist seals. This model is also intended for more constant wear. The vessel suffered a catastrophic failure of the factory refrigeration system, which resulted in a propulsion system failure and a loss of electrical power. The loss of electrical power prevented the operation of the wastewater discharge pumps on the factory deck. Seawater entered the factory deck through partially obstructed overboard discharge valves, which were not fitted with anti-syphon loops. Seawater entering the factory deck caused the vessel to heel to starboard and trim by the stern. The engine room door on the factory deck and other weather and watertight closings at the after end of the weather deck were not effectively sealed, permitting the accumulated water to downflood. The accumulated water downflooded from the factory deck over the coaming of the open engine room door into the engine room. The heel to starboard and the settling by the stern increased until the vessel finally sank.Findings as to Causes and Contributing Factors The vessel suffered a catastrophic failure of the factory refrigeration system, which resulted in a propulsion system failure and a loss of electrical power. The loss of electrical power prevented the operation of the wastewater discharge pumps on the factory deck. Seawater entered the factory deck through partially obstructed overboard discharge valves, which were not fitted with anti-syphon loops. Seawater entering the factory deck caused the vessel to heel to starboard and trim by the stern. The engine room door on the factory deck and other weather and watertight closings at the after end of the weather deck were not effectively sealed, permitting the accumulated water to downflood. The accumulated water downflooded from the factory deck over the coaming of the open engine room door into the engine room. The heel to starboard and the settling by the stern increased until the vessel finally sank. There was a 14 hour delay in notifying the appropriate authorities of the vessel's situation. Although not a causal factor, the vessel's fuel oil distribution at the time of the occurrence did not comply with the restrictions in the vessel's approved Trim and Stability Booklet.Findings as to Risk There was a 14 hour delay in notifying the appropriate authorities of the vessel's situation. Although not a causal factor, the vessel's fuel oil distribution at the time of the occurrence did not comply with the restrictions in the vessel's approved Trim and Stability Booklet. Several problems were identified with the immersion suits during the abandoning of the vessel.Other Findings Several problems were identified with the immersion suits during the abandoning of the vessel. Safety Action Action Taken Delay in calling for assistance A Marine Safety Information letter (No06/02) was sent to the Canadian Coast Guard, and copied to Transport Canada, on 07August2002. Department of Fisheries and Oceans The Search and Rescue Branch of the Canadian Coast Guard reviewed and updated the 2003edition of the Radio Aids to Marine Navigation (RAMN), page4-33 Alerting the Search and Rescue Authorities (Marine Safety Circular No892), which describes why early alerting is necessary and also gives operational guidance for masters of vessels in distress or urgency situations. The 2003edition of Notice to Mariners, Section28, was also reviewed and updated by the addition of a new paragraph, titled Importance of Early Notification of a Potential Distress. In response, Transport Canada (TC) drew attention to Ship Safety Bulletin No06/2001, issued on 08August2001, titled Global Maritime Distress and Safety System and Guidance on Important Operational Procedures. This issue was to inform mariners of a number of radio operational procedures essential for safety. Annex3 of the Bulletin (Section1) clearly identifies the need for an early alerting of search and rescue and that (Section2) it is essential to enable shore-based facilities to respond without delay to any situation which constitutes, or has the potential to constitute, a danger to life. Time lost in the initial stages of an incident may be crucial to its eventual outcome. TC advised that the CCG RAMN, Annual Edition2001, Part4 - General Procedures, re-iterates the need for an early alerting, and to help ensure maximum dissemination and availability of this information, the material will be published in the2002 edition. This publication is required to be carried aboard all vessels fitted with a ship radio station. Anti-syphon Loop A Marine Safety Information letter (No07/02) was sent to Transport Canada, on 07August2002. In their initial response, TC did not specifically address the anti-syphon issue. However, they advise that they are developing a new construction standard for fishing vessels which would include piping systems and overboard discharges. Discussion on the new standard will take place at the Canadian Marine Advisory Council meeting in May2003, but specific aspects will be addressed later during the Regulatory Reform Process when proposals regarding anti-syphon arrangements will be tabled. Immersion Suits The manufacturers of the OC4001 and OC8001 immersion suits improved the construction of the OC4001 suits to permit more constant use. Stored immersion suits require periodic on-board visual inspections. It is recommended that, after each emergency use, the suit be subjected to an air pressure test. The United States has submitted a recommendation to the International Maritime Organization for guidelines respecting periodic testing of stored immersion suits. Air pressure tests, at intervals not exceeding 3years, have been suggested or more frequently for suits over ten years of age. The Board will monitor this initiative. This report concludes the TSB's investigation into this occurrence. Consequently, the Board authorized the release of this report on 24February2003.